riders on the storm - home, the social market … · riders on the storm contents acknowledgements...

Nida Broughton, Onyinye Ezeyi, Claudia Hupkau

Britain’s middle income households since 2007

RIDERS ON THE STORM

THE SO

CIA

L MA

RK

ET FOU

ND

ATIO

NR

IDER

S ON

THE STO

RM

: BR

ITAIN

’S MID

DLE IN

CO

ME H

OU

SEHO

LDS SIN

CE 2007

N

IDA

BR

OU

GH

TON

, ON

YIN

YE EZ

EYI, C

LAU

DIA

HU

PKA

U

Copyright © Social Market Foundation, 2014

ISBN: 978-1-904899-89-1

£10.00

Nida Broughton, Onyinye Ezeyi and Claudia Hupkau

Britain’s middle income households since 2007

RIDERS ON THE STORM

FIRST PUBLISHED BY The Social Market Foundation, April 2014 ISBN: 978-1-904899-89-1

11 Tufton Street, London SW1P 3QBCopyright © The Social Market Foundation, 2014The moral right of the authors has been asserted. All rights reserved. Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the prior written permission of both the copyright owner and the publisher of this book.

THE SOCIAL MARKET FOUNDATIONThe Foundation’s main activity is to commission and publish original papers by independent academic and other experts on key topics in the economic and social fields, with a view to stimulating public discussion on the performance of markets and the social framework within which they operate.

The Foundation is a registered charity and a company limited by guarantee. It is independent of any political party or group and is funded predominantly through sponsorship of research and public policy debates. The Foundation takes complete responsibility for the views expressed in this publication, and these do not necessarily reflect the views of the sponsor.

CHAIRMANMary Ann Sieghart

MEMBERS OF THE BOARDDaniel FranklinLord John HuttonGraham MatherSir Brian Pomeroy cbeBaroness Gillian ShephardNicola HorlickProfessor Alison Wolf

DIRECTOREmran Mian

DESIGN AND PRODUCTIONSoapbox www.soapbox.co.uk

PRINTED BY Repropointwww.repropoint.com

RIDERS ON THE STORM

CONTENTS

Acknowledgements 4

About the authors 5

Executive summary 6

1. WHY WORRY ABOUT MIDDLE INCOME HOUSEHOLDS? 10

2. WHO IS IN THE MIDDLE, AND HOW HAS THE MIDDLE CHANGED? 13What the middle looks like 13

How the middle has changed since 2007 15

3. COPING WITH THE DOWNTURN 20Keeping money flowing in 20

Keeping costs under control 21

The future squeeze 26

“Are you better off than you were four years ago?” 28

4. HELPING MIDDLE INCOME HOUSEHOLDS 30Comparing and switching 32

Giving people access to the right information 33

Liberating data 34

Just pushing data is not enough 36

ANNEX: METHODOLOGY 39Overview 39

Analysis of Understanding Society and the

British Household Panel Study 40

Analysis of the Family Resources Survey 42

Analysis of the Living Costs and Food Survey 43

SOCIAL MARKET FOUNDATION

4

ACKNOWLEDGEMENTS

The publication of this report has been made possible by the

generous support of Santander.

Our Families Panel, established to advise on the SMF’s

programme of work on middle income families, provided valuable

thoughts and insights on our initial data findings. Thanks are due

to Jonathan Akerman, Chris Bose, Elizabeth Duff, Duncan Hames,

Adrian Harvey, Matt Oakley and Sue Willis. We would also like to

thank Which? for providing data extracts from their Consumer

Insight Tracker.

At the SMF, we would like to thank Emran Mian and Nigel

Keohane for intellectual input throughout the project.

RIDERS ON THE STORM

5

ABOUT THE AUTHORS

NIDA BROUGHTON

Nida Broughton is Chief Economist at the SMF, where she has led

research on skills policy, entrepreneurship and analysis of public

spending, as well as undertaking research on a range of public policy

areas including healthcare, education and housing. Nida previously

worked at the House of Commons, where she advised MPs and

committees on a broad range of economic issues, and in particular,

on financial services. Nida also has in-depth expertise in consumer

regulation, competition policy and behavioural economics, gained

whilst working at Ofcom, the UK regulator and competition authority

for communications markets. She has an MA (Cantab) in Economics

from Cambridge University and an MSc in Economics from Birkbeck

College, University of London.

ONYINYE EZEYI

Onyinye Ezeyi is a Research Intern at the Social Market Foundation.

She recently completed an MSc in Development Economics from

the University of Sussex after graduating with a First in BSc (Hons)

Economics with Sociology from Loughborough University in 2011.

Onyinye’s interests lie in the use of econometric methodologies in

social research and has worked on projects in migration and health.

She plans to study for a PhD in Economics.

CLAUDIA HUPKAU

Claudia Hupkau is a Research Fellow for the Social Market

Foundation. She is currently completing a PhD in Economics at the

Université Catholique de Louvain (UCL), Belgium, and is a member

of the Centre for Operations Research and Economics, Belgium. She

is part of the research group ‘Poverty, Resource Equality and Social

Policy’ at UCL and has worked on the behavioural economics of

poverty and education.


6

EXECUTIVE SUMMARY

In the years since the financial crisis, households across the income

distribution have seen a combination of wage stagnation and a rise

in the cost of living. Working age households in particular have been

adversely affected by real declines in their main source of income.

Much analysis has focused on lower income households. But

more recently, the so-called “cost of living crisis” has pushed the

political focus onto the “squeezed middle”. Households in the

middle and fourth quintile of the income distribution have annual

incomes between around £26,100 and £63,000. They receive some

cash benefits, but given the pressure to reduce public spending,

they are unlikely to be prime candidates for more state transfers,

leaving them more reliant on their own resources in coping with

the recent economic climate.

So how well have they fared? Is it indeed the case that they

have been squeezed? How successfully have they coped with

rising prices whilst incomes have stagnated?

The SMF’s analysis draws on the British Household Panel and

Understanding Society surveys to follow specific households through

time, from the brink of the downturn in 2007–08 to 2011–12, the latest

wave of available data. We complement this with data from other

national surveys to look at specific types of spending in more detail.

What is most striking is that middle income households today

are not the same households that were in the middle going into

the downturn. The next most striking finding is the very diverse

experiences of the middle. Of those who were in the middle in

2007–08:

• 6% had dropped into the bottom fifth of the distribution by

2011–12

RIDERS ON THE STORM

7

• Just over 10% had climbed up to the top fifth of the

distribution by 2011–12

• Overall the households that climbed up the distribution

outnumbered those that fell.

Correspondingly, today’s middle income households have

seen a diverse range of journeys to where they are now. Of those

households in the middle quintile in 2011–12:

• One in 10 were in the bottom fifth of the income distribution

four years earlier

• One in five were in the second to bottom quintile four years

earlier

• 17% were previously higher up the distribution and dropped down.

Overall, this combination of climbers and fallers meant that

those in the middle by 2011–12 saw their average real income

remain roughly the same since 2007–08. Meanwhile, the quintile

just above – the fourth quintile – saw a real increase. In contrast,

those who were in the bottom two quintiles by 2011–12 had seen

falls in real income.

Moving up into and remaining in the middle has been made

possible through higher employment levels. Households in the

middle today are more likely to have two earners rather than one

compared to 2007–08. They have managed spending on food –

switching to cheaper items – to avoid the 25% increase in food

prices over the period. Childcare is still problematic for many, with

a heavy reliance on grandparents compared to those on higher

incomes. But they have also been helped by low interest rates,

which have kept housing costs down among the large number of

home owners in this category.

For many middle income households, there are rising costs to

come when interest rates go up again. With competing pressures


8

on public spending, there is less scope for more state help. But

this does not mean that nothing can be done. One area of policy

where these households can be helped to do more with less is

in consumer markets. Those in the middle today have displayed

significant resilience thus far in coping with the downturn. But

there are markets where more needs to be done to help these

households help themselves in securing better value for money.

Substantial progress has already been made on making

switching processes easier. Now, the bigger gains are likely to come

from helping consumers easily compare products and services.

This means ensuring there is sufficient information easily available,

that product features and pricing are transparent, and that it is easy

to use information that does exist to calculate likely savings from

switching.

For middle income families in particular, managing childcare

costs is difficult. This is not helped by the fact that childcare is one

area where information that allows parents to compare providers is

very sparse. Whilst there is some information on quality, comparing

prices and value for money is much harder. Helping parents access

this information will help them make better choices about what

childcare is best for them.

There is also a need to ensure that consumers have access to

the right information and that product features and pricing are

transparent. As interest rates rise again, the middle is likely to see

further pressure on their spending. Last year saw rises in tracker

mortgage rates despite the Bank of England base rate remaining

the same. It is vital that there is sufficient clarity from banks and

regulators as to what consumers should expect from different

types of products so that they can make informed choices about

how to manage the future rise in mortgage costs. Measures such as

the implementation of the Financial Conduct Authority’s Mortgage

Market Review will help with this.

RIDERS ON THE STORM

9

Finally, making information easy to access and use to calculate

savings is vital. Consumers could greatly benefit if they had better

access to their own data to compare prices and providers, as is

happening through the midata initiative. Even better would be the

development of services that allow consumers to compare how much

they are spending compared to the average, as an indicator of the

level of savings possible to achieve by switching. Other products and

services – beyond energy, mobile phones, current accounts and credit

cards – should be considered as part of the midata strategy. These

include telecoms and broadband more generally; savings products

and groceries. Government, regulators and industry will need to

ensure that major players in the industry act together, whether this is

achieved through voluntary agreement or compulsion.

But pushing data alone will not always be enough. Even when

the right information is available and easily comparable, evidence

suggests that prompts may be needed to encouraging switching.

New, innovative apps that scan the market and alert consumers to

new deals may be developed. Where this does not happen, other

measures should be considered. The model of renewal notices

in the car insurance market, which consumers say prompts them

to shop around, could be considered for other sectors, and if

combined with easy to use usage data, could be very powerful.

Given that Government is less than halfway through its

spending cuts programme, there will be many competing

pressures on public expenditure. But this does not mean that

middle income households cannot be helped. These households

have shown remarkable resilience so far, demonstrating that

under the right circumstances, and with the right information to

hand they can manage their finances and cope with being under

pressure. There should be greater focus on equipping the middle

to negotiate consumer markets where it is harder to secure better

value for money. Making this happen is likely to require action from

Government, regulators and industry.


10

1. WHY WORRY ABOUT MIDDLE INCOME HOUSEHOLDS?

Households across the income distribution have seen a combination

of wage stagnation and rises in the cost of living since 2008. The

period 2010 to 2013 in particular is the longest sustained period of

falling real wages in the UK on record.1

Working age households have borne the brunt, and specifically

have been adversely affected by the fall in real wages. Among these

working age households, those in the middle fifth receive less state

help in the form of benefits compared to lower quintiles, although

they do receive some, as shown in Chart 1.1. Instead, they have been

more reliant on their own resources to cope with the downturn.

Chart 1.1: Benefits and taxes as a proportion of original income

0%

20%

-20%

40%

-40%

60%

-60%

80%

-80%

100%

Cash benefits

Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5

Direct taxes Indirect taxes

Office for National Statistics, “The effect of taxes and benefits on household income, 2011–12”, released 10 July 2013

Note: Non-retired households, quintile groups ranked on equalised disposable income

The middle, and to some extent the fourth, quintile is of specific

interest. Unlike the very top quintile, they are less likely to have

been able to accumulate savings to help negotiate the downturn.

1 Office for National Statistics, “An examination of falling real wages, 2010–2013”, 31 January 2014, http://www.

ons.gov.uk/ons/dcp171766_351467.pdf

RIDERS ON THE STORM

11

The middle is also subject to broader economic and long-term

pressures driven by globalisation and technological change. There

has been a “hollowing out” effect, whereby the proportion of jobs in

low wage and high wage occupations has risen and the proportion

in middle wage occupations has fallen. Whilst this effect is likely to

be relatively small on a year-to-year basis, evidence suggests that

the recession accelerated the decline in middle wage jobs.2

This report examines the reality of the “squeezed middle”:

looking at who the middle are, where they were in 2007–08 and

how they coped with the downturn. To do this, we use longitudinal

data – from the British Household Panel (BHPS) and Understanding

Society (US) surveys,3 that follow specific households over time.

We compare income and spending in two years: 2007–08, on the

brink of the downturn, and 2011–12, the latest data available. Where

longitudinal data is unavailable, we complement this analysis with

data from the Family Resources Survey (FRS)4 and the Living Costs

and Food Survey (LCFS).5 Full detail on methodology and sources

are included in the Annex to this report.

We find that those in the middle by 2011–12 were precisely

those families that had weathered the downturn relatively well.

Employment for these households held up – and sometimes

increased – and they were able to manage spending to minimise

the effect of rising prices. But many of these households will see

rising costs in the future; when interest rates rise, mortgage costs,

which have been kept low, are likely to rise again. The final section

2 Paul Sissons, The hourglass and the escalator (London: The Work Foundation, 2011)

3 University of Essex. Institute for Social and Economic Research and National Centre for Social Research,

Understanding Society: Wave 3, 2011–2012 Colchester, Essex: UK Data Archive SN: 6614; British Household Panel

Survey: Wave 17, 2007–08 Colchester, Essex: UK Data Archive SN: 5151.

4 Department for Work and Pensions, National Centre for Social Research and Office for National Statistics.

Social and Vital Statistics Division, Family Resources Survey, 2007–08 and 2010–2011. Colchester, Essex: UK Data

Archive, SN: 7085 , http://dx.doi.org/10.5255/UKDA-SN-7085-1

5 Office for National Statistics and Department for Environment, Food and Rural Affairs, Living Costs and Food Survey,

2011 Colchester, Essex: UK Data Archive, August 2013. SN: 7272 , http://dx.doi.org/10.5255/UKDA-SN-7272-2


12

of this report looks at what policymakers can do to ease financial

pressures in the future.

The report is set out as follows:

• Chapter 2 sets out what the middle looks like, and how it has

changed since 2007–08, looking at trends in income.

• Chapter 3 shows how middle income households coped with

the downturn, by keeping money flowing in and managing

spending. It also looks at the effect of the downturn on their

well-being and prospects for the future.

• Chapter 4 discusses the main policy lever available to help

middle income households given the tightening in public

finances: making consumer markets work better.

RIDERS ON THE STORM

13

2. WHO IS IN THE MIDDLE, AND HOW HAS THE MIDDLE CHANGED?

This chapter sets the scene, showing what the middle looks like in

terms of employment and occupations, and how the middle has

changed since the downturn.

In our analysis, we focus on working age middle income

households, defined as the middle fifth or middle quintile. We also

look at the quintile just above, quintile 4. More specifically, we rank

households by their original income, before taxes and transfers,

from the top income households in quintile 5 to the bottom

income households in quintile 1. The purpose of using the original

income measure as opposed to other measures of income is to

better understand the underlying trends and changes that middle

income families underwent.

In our dataset in 2011–12, middle quintile households had

an average annual original income ranging between £26,100

and £41,200. On average, original income of these households

was £33,600. Those in the fourth quintile had an average annual

income of £51,100, with a range of between £41,200 and £63,000.

This section sets out the key characteristics of middle income

households, based on analysis of the BHPS and US.

WHAT THE MIDDLE LOOKS LIKE

The majority of middle income households – 91% – had at least

one person in employment in 2011–12. In addition, around half

of these households had two wage earners. Looking at the

proportion of doubler-earner households across the income

distribution in Chart 2.1, it is clear that having dual income sources

from two earners is a key factor that helps households reach the

higher end of the distribution. Where households in the middle

fifth differ from the quintile just below them is in the much higher


14

rate of double-earner households: 50% compared to just 21% in

the second quintile.

In contrast, the fourth quintile has even higher rates of double-

earner households compared to the middle – 64% . This is likely

to be at least partly due to the fact that whilst middle and fourth

quintile households are as likely as each other to have children,

the middle quintile are slightly more likely to have children under

10. The presence of younger children may make it harder for

households in the middle to have two adults working, and as we

will see later on, is likely to create particular childcare challenges for

these households.

Chart 2.1: Proportion of households with individuals in employment

100%90%80%70%60%50%40%30%20%10%0%

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

Proportion of households with two earnersProportion of households with at least one person employed

SMF analysis of Understanding Society, 2011–12, Wave 3

In terms of occupations, the middle quintile has much in

common with the quintiles below. A substantial number of

individuals in the middle and fourth quintiles are in occupations

likely to be low-wage. Half of individuals in middle income

households are in the four lowest skill occupations, as shown in

Chart 2.2. As such, those in the middle are strongly affected by

labour market conditions in the lower wage occupations.

RIDERS ON THE STORM

15

Chart 2.2: Proportion of individuals split by occupation

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

Managers, directors and senior officialsProfessional occupationsAssociate professional and technical occupationsAdministrative and secretarial occupationsSkilled trades occupations

Caring, leisure and other service occupationsSales and customer service occupationsProcess, plant and machine operativesElementary occupations

100%80%60%40%20%0%


HOW THE MIDDLE HAS CHANGED SINCE 2007

Real wages have seen a substantial decline following the crisis and

downturn. But in the last few years, different households have had

varying experiences of the downturn. This means that in addition to the

overall stagnation of wages, the shape of British society has changed.

Chart 2.3 shows the UK’s income distribution for working age

households in 2007–08 and 2011–12. While at the very top the

distribution has hardly changed, the middle and bottom have

shifted to the left, meaning that real incomes have decreased

for all but the top deciles. As of 2011–12, the average household

in the middle quintile had real income of 13% per cent less than

the middle quintile in 2007–08 – before taxes and benefits, as

real earnings fell.6 Because incomes have become more skewed

6 The ONS has calculated that disposable income of non-retired middle quintile fell by 6.4%. Disposable

income is the amount of money that households have available for spending and saving after direct taxes

(such as income tax and council tax) have been accounted for. It includes earnings from employment, private

pensions and investments as well as cash benefits provided by the state; Office for National Statistics, “Middle

Income Households, 1977–2011/12”, 2 December 2013, http://www.ons.gov.uk/ons/dcp171776_341133.pdf


16

towards the bottom, it is now “easier” to get into the middle fifth in

terms of absolute income required.

Chart 2.3: UK income distribution in 2007–08 and 2011–12

0

.02

.04

.06

.08

Frac

tion

0 5,000 10,000 15,000

0

.02

.04

.06

.08

Frac

tion

0 5,000 10,000 15,000

Monthly household income before benefits

Monthly household income before benefits

2007–08

2011–12

SMF analysis of Family Resources Survey 2007–08 and 2011–12. Restricted to incomes between 0–15000 to

exclude outliers. 2011–12 prices

This has a number of implications for attempting to understand

how the middle has changed. The fall of many people to the

bottom of the distribution means that households in 2011–12

needed a lower income to be classified as in the middle. The gap

between low and middle income has narrowed.

If all household incomes had shifted downward by the same

amount, there would be no difference between the households

that were in the middle in 2007–08 and those that were in the

middle in 2011–12. But this was not the case. Many of the households

RIDERS ON THE STORM

17

that previously formed part of the middle have themselves moved

elsewhere in the distribution, as shown in Chart 2.4. The middle of

2007–08 is not the same as the middle of 2011–12.

Chart 2.4: Where the 2007–08 middle and fourth quintiles went

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Quintile 1, 2011–12Quintile 4, 2011–12


Quintile 3, 2011–12

Quintile 3, 2007–08 Quintile 4, 2007–08

SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3

In fact, one in 10 of those classified as being in the middle in

2007–08 made it all the way to the top quintile four years later,

and 31% made it into the fourth quintile. 6% fell into the bottom

quintile, but overall, the “climbers” from the middle income quintile

outnumbered the “fallers”.

What of the households that were in the middle coming out

of the downturn? The so-called “squeezed middle” of 2011–12 is

a mixed group. Over a quarter were made up of households that

used to be in the top two quintiles. But there has been remarkable

upward movement too, with one in 10 of them coming from the

bottom quintile, as shown in Chart 2.5.


18

Chart 2.5: Where the 2011–12 middle and fourth quintiles came from



100%90%80%70%60%50%40%30%20%10%0%


Quintile 2, 2007–08 Quintile 3, 2007–08Quintile 5, 2007–08


Overall, this combination of climbers and fallers means that

those in the middle as of 2011–12 saw – on average – no significant

change in income. It has become in a sense easier to enter the

middle quintile: the income threshold is lower than it used to be.

This means that although the climbers outnumber the fallers, the

climbers did not require as big an income rise as they would have

previously needed to be considered middle income. But as we shall

see in the next chapter, employment played a significant part in

helping households climb into and remain in the middle.

At first this may seem counter-intuitive given the fall in real

income between the middle quintile in 2007–08 compared to the

middle quintile in 2011–12. But this real income fall is a snapshot

analysis, that does not reflect the fact that many of those in the

middle by 2011–12 were not in the middle four years previously: they

were the downturn’s success stories, moving up to the middle from

the bottom quintiles. Tracking actual households over time allows

us to better understand where the middle of today comes from,

compared to a snapshot analysis. The households in the middle by

2011–12 were often the households that had seen improvements in

their financial situation compared to the average.

RIDERS ON THE STORM

19

Chart 2.6: Original monthly income before taxes and benefits

(2011–12 prices)

£5,000

£4,500

£4,000

£3,500

£3,000

£2,500

£2,000

£1,500

£500

£0

£1,000

2007–08 2011–12

Quintile 3 in 2011–12 Quintile 4 in 2011–12

£2,961

£3,909

£2,817

£4,253


Error bars represent 95% confidence intervals

Quintiles are defined based on the 2011–12 income distribution. The chart compares the incomes of the same

group of households in 2007–08 and 2011–12.

The rest of this report looks at how those in the middle today

have coped with the downturn – how they have managed to

maintain their incomes and keep a lid on spending during a time

of rising prices.


20

3. COPING WITH THE DOWNTURN

This chapter looks at how those in the middle today coped with the

downturn, and how this affected their well-being.

KEEPING MONEY FLOWING IN

Those who were younger when the recession hit were more likely

to have moved up to the middle quintile by 2011–12 – most likely

due to the fact that they were on low wages at the start of their

careers, and saw income increases as they progressed on. But this

“career lifecycle” effect explains the smaller part of the change in

households’ relative positions.

Employment – and especially employment of a second

household member – is the main driver affecting a household’s

position in the income distribution. Many middle income

households in 2007–08 which lost a second earner fell down the

income scale. In the group of households that climbed up from the

bottom two quintiles into the middle, the proportion of double-

earner households rose from 35% to 49%. In contrast the group

that fell down from the top quintiles into the middle saw their

proportion of double earner households drop from 72% to 62%.

Overall, one in 20 of those in the middle by 2011–12 had

become a double-earner household over the preceding four years.

The majority of these new second earners – two thirds – had full

time jobs.

RIDERS ON THE STORM

21

Chart 3.1: Proportion of households with two earners

70%60%50%40%30%20%10%0%

Quintile 1in 2011–12




2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12



KEEPING COSTS UNDER CONTROL

EssentialsMiddle income households spend around 30% of their gross income

on essentials – food, mortgage and rent, and energy. Mortgage and

rent is the largest of these, and varies substantially across the UK.

Those on middle incomes in the North East and North West spend

about 15% of their gross income on these housing costs; in London,

this rises to 22%.

Chart 3.2: Spending on essentials as a proportion of gross income, 2011–12

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

Food Energy Mortgage/rent

60%40%20%0% 10% 30% 50%



22

But during the downturn, middle income households have been

protected from rising housing costs. Over half of these households

own a house with a mortgage, and 15% own outright. Historically

low interest rates mean that their housing costs have either remained

steady or declined since 2007. The Bank of England base rate was over

5% in 2007; in contrast it has been 0.5% since 2009.

Where households have felt the pinch is on food. Food

prices rose by 25% from 2007 to 2011 and the cost of meals from

restaurants and cafes rose by 16%.7 However, middle income

households cut back on expensive items and switched to cheaper

ones, keeping the spending rise across groceries and meals to 10%

– below inflation. Households that fell down from the top quintiles

kept the food spending rise even lower.

They have also felt the pinch on energy. Whilst we cannot

compare energy spend directly because of temperature differences

in the two years we look at, we do know that the price of electricity,

gas and fuel rose by 35% from 2007 to 2011, on the CPI measure

of inflation. Data from the ONS (not specific to middle income

households) suggests that spending increased sharply over the

period 2007 to 2009. After 2010 total spending remained steady

– but partly due to reduced consumption as a result of milder

winters. More generally, over the past decade, households have

been reducing energy consumption through energy efficiency

measures and simply using less in response to higher bills.8

A similar story is apparent for transport spending, where costs

have increased by around 24% over the period we look at, on the CPI

measure of inflation. ONS analysis of general household spending

7 ONS, “Consumer Price Inflation Reference Tables”, January 2014, http://www.ons.gov.uk/ons/rel/cpi/

consumer-price-indices/january-2014/consumer-price-inflation-reference-tables.xls.

8 Office for National Statistics, “Household energy spending in the UK, 2002–2012”, 3 March 2014, http://

www.ons.gov.uk/ons/rel/household-income/expenditure-on-household-fuels/2002---2012/full-report--

household-energy-spending-in-the-uk--2002--2012.html

RIDERS ON THE STORM

23

finds that spending on transport has seen a general decline over

the past decade. This is likely to be at least partly due to consumers

responding to higher fuel prices by cutting car use and switching

to public transport; and switching to more energy efficient cars.9

In many middle quintile households, two adults go out to

work. But with less income than the top quintile, middle income

households are the most reliant on informal sources of childcare such

as grandparents, as shown in Chart 3.3. Those on higher incomes, in

the top quintile, are in contrast able to rely more on formal childcare.10

Chart 3.3: Proportion of households using grandparents for childcare

60%

50%

40%

30%

20%

10%

0%

2007–08 2011–12

Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5

SMF analysis of Family Resources Survey 2007–08 and 2011–12

Whilst the proportion that rely on grandparents has slightly fallen

since 2007, those who rely on informal care have been using more

of it, at least in part due to rising childcare costs. The effect is most

clear for those with children under the age of two. Charts 3.4 and

3.5 show trends in spending and hours used. Whilst parents in the

middle quintile have barely increased the number of paid-for hours

9 Office for National Statistics, “Family Spending”, 11 December 2013, http://www.ons.gov.uk/ons/rel/family-

spending/family-spending/2013-edition/index.html

10 Note for childcare costs, we rely on FRS data. FRS does not follow households longitudinally. Comparisons

are therefore made between quintiles as defined in 2007–08 against quintiles as defined in 2011–12.


24

of care they use, they were paying 22% more for childcare in 2011–

12 compared to 2007–08, in real terms. And research by the Family

and Childcare Trust show that nursery costs for this age group have

continued to rise.11

Chart 3.4: Change in spending for children aged under two

70

60

50

40

30

20

10

0

2007–08 2011–12

Quintile 1 Quintile 2

£ pe

r wee

k pe

r chi

ld

Quintile 3 Quintile 4 Quintile 5

SMF analysis of Family Resources Survey 2007–08 and 2011–12.

Chart 3.5: Change in hours of childcare used for children aged under

two, 2007–08 to 2011–12

5.0

4.0

3.0

2.0

1.0

0.0

-1.0

-2.0

Free hours Paid-for hours

Quintile 1

Cha

nge

in h

ours

per

wee

k pe

r chi

ld

Quintile 2 Quintile 3 Quintile 4 Quintile 5

SMF analysis of Family Resources Survey 2007–08 and 2011–12.

11 Jill Rutter and Katherine Stocker, Childcare costs survey 2014 (London: Family and Childcare Trust, 2014)

RIDERS ON THE STORM

25

Non-essentials12 Households have reduced spending on leisure goods but not leisure

services. As shown in Chart 3.6, the drop in leisure goods spending

has been especially sharp for the middle and fourth quintiles. In the

case of leisure goods – which include products such as television

sets, phones, newspapers, books, games consoles and DVDs –

households have been helped by the fact that inflation has been

relatively muted. Prices for audio-visual equipment for instance

fell by more than 50% since 2007, general recreational items saw a

price rise of only 1.5% over the period from 2007 to 2011, and other

leisure goods saw a level of inflation in line with the general price

increase of about 14% during that period.13

In contrast, leisure services – which include sports clubs,

cinemas, live entertainment and leisure subscriptions and holidays

– generally saw inflation above or in line with the average, making

it harder to cut back, as shown in Chart 3.7.

Chart 3.6: Spending per week on leisure goods (2011 prices)

£40£35£30£25£20£15£10£5£0

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

2011 2007

SMF analysis of Living Costs and Food Survey 2007 and 2011

12 Note for leisure costs, we rely on LCFS data. LCFS does not follow households longitudinally. Comparisons

are therefore made between quintiles as defined in 2007–08 against quintiles as defined in 2011–12.

13 ONS, “Consumer Price Inflation Reference Tables”, January 2014, http://www.ons.gov.uk/ons/rel/cpi/

consumer-price-indices/january-2014/consumer-price-inflation-reference-tables.xls.


26

Chart 3.7: Spending per week on leisure services (2011 prices)

£40 £60 £80 £100 £120 £140 £160 £180£20£0

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

2011 2007

SMF analysis of Living Costs and Food Survey 2007 and 2011

THE FUTURE SQUEEZE

Overall, those in the middle today have been remarkably resilient in

changing their spending habits to suit their means. But in a sense

they are yet to be fully tested. Over two-thirds of middle quintile

households are homeowners, and just over half have a mortgage.

The high proportion of homeowners with a mortgage has helped

middle income households so far; but this also means they will face

rising costs when interest rates rise again.

A key determinant of the impact of this squeeze will be how

the Bank of England raises interest rates, and whether rates rise in

tandem with income growth. Bank of England analysis recently

showed that a rise in interest rates of 2.5 percentage points would

increase the proportion of households with high repayments

compared to incomes by 50% if incomes rise by 10%, and double

the proportion if incomes remained unchanged.14 Within the

middle income group, this squeeze is likely to affect 30–50 year

14 Philip Bunn, May Rostom, Silvia Domit, Nicoal Worrow, Laura Piscitelli, The financial position of British

households, (London: Bank of England, 2013)

RIDERS ON THE STORM

27

olds the most. This age group is more likely than older cohorts to

own a home with a mortgage rather than own outright.

Even during this period of low interest rates, over one in 10 of

middle income households in 2011–12 reported being behind on

rent or mortgage at least once in the past year, with households

with children under more pressure, as shown in Chart 3.8. Many

of these households are likely to have a significant amount left to

repay on their mortgages. Whilst data is not available on the middle

quintile specifically, across all households, the proportion with high

loan to value ratios was higher in 2013 compared to before the

financial crisis. Just under a fifth have a loan to value ratio of over

75% on their outstanding debt.15

Chart 3.8: Proportion of households behind on rent or mortgage,

2011–12

5% 10% 15% 20% 25%0%

Quintile 5

Quintile 4

Quintile 3

Quintile 2

Quintile 1

Households with children All households


Looking more widely, 8% of middle quintile households said

they defaulted on a loan, bill, mortgage or rent in the past month.16

Middle income households have also – in part – been relying on

15 Philip Bunn, May Rostom, Silvia Domit, Nicoal Worrow, Laura Piscitelli, The financial position of British households

16 Which? “Consumer Insight Tracker”, February 2014


28

credit to help fill gaps. 13% of middle quintile households took out

a loan or credit card, borrowed from friends or family, or used an

authorised overdraft in the last month.17

“ARE YOU BETTER OFF THAN YOU WERE FOUR YEARS AGO?”

Almost 60% of middle income households say they are “comfortably

off” or “doing alright”. Having weathered the downturn reasonably

well compared to other households, they are – on the face of it

– fairly comfortable. But they certainly do not feel better off than

they were back in 2007, as shown in Chart 3.9.

Chart 3.9: Proportion of households answering “How well are you

managing financially?”

0%

10%

20%

30%

40%

50%

60%

70%

80%

Living comfortably Doing alright

Quintile 3 in 2011–12 Quintile 4 in 2011–12

2011–12 2011–122007–08 2007–08


Quintiles are defined based on the 2011–12 income distribution. The chart compares responses of the same

group of households in 2007–08 and 2011–12.

They are less satisfied with the amount of leisure time they

have, and less satisfied with life overall. This pattern is hardly unique

to the middle, as shown in Charts 3.10 and 3.11. But interestingly,

our analysis shows that life satisfaction patterns do not change

17 Which? “Consumer Insight Tracker”, February 2014

RIDERS ON THE STORM

29

substantially depending on how middle income households’

relative position in the income distribution shifted, indicating

that feeling better off is not necessarily strongly driven by actual

changes in income. It may instead be more related to prospects

for the future: which in 2011–12 were likely to be looking gloomy

compared to 2007–08.

Chart 3.10: Life satisfaction (proportion of individuals satisfied minus

proportion of individuals dissatisfied)

70% 80%60%50%40%30%20%10%0%





2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12



Chart 3.11: Satisfaction with amount of leisure time (proportion of

individuals satisfied minus proportion of individuals dissatisfied)

35% 40%30%25%20%15%10%5%0%





2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12

2007–08

2011–12




30

4. HELPING MIDDLE INCOME HOUSEHOLDS

Middle income families do not feel better off even though – in

relative terms – they are generally the ones that have weathered

the downturn well. They have kept their jobs and cut spending by

switching to cheaper items and cutting back. But some of squeeze

may be still to come. Depending on the recovery of real wages,

many of these households could find it harder to manage their

finances in the years ahead.

Despite widespread focus on the “squeezed middle”, in reality,

these households are unlikely to see substantial additional state

help in the form of tax breaks and credits, given the competing

pressures on Government spending. But these households can be

helped to do more of what they have already done well: shopping

around, comparing prices and making their money go further.

In markets where there is significant choice and competition,

such as food, consumers have been able to manage price increases

by changing what they buy and switching. In other markets, this is

harder to do. In energy markets, despite bill increases, consumer

switching has actually been falling since 2009.18 Whilst consumers

have been saving money by using less energy, switching appears –

in comparison – to be an untapped source for cutting costs.

Our data shows substantial variation in the amount spent

per month on energy across households within the five quintiles,

as shown in Chart 4.1. Variation is in part due to differences in

consumption. But other statistics suggest that some households

are not getting the best deal possible. Data from DECC shows

that consumers could be paying anywhere between 12.79 pence

per kilowatt hour to 18.69 pence per kilowatt hour for electricity

depending on which tariff they pick and the payment method

18 Oliver Finlay, Illuminating the energy market (London: Social Market Foundation, 2013)

RIDERS ON THE STORM

31

they choose.19 Even within regions, bill sizes can vary substantially

depending on tariff.

Chart 4.1: Range in spending per week on energy across the

quintiles, 2011

£0

£10

£20

£30

£40

£50Quintile 1 Quintile 2

Quintile 4 Quintile 5

Quintile 3

£0

£10

£20

£30

£40

£50

SMF analysis of Living Costs and Food Survey 2011.

Excludes outside values, based on ONS fuel, light and power variable

Financial services is another area where consumers are

foregoing savings by not shopping around for the best deal.

Around two in five savers earn interest of just 0.5 per cent or less,

and one in five receive just 0.1 per cent or less a year.20

So what can be done? In some markets, even where consumer

engagement and switching is high, the market structure works to

constrain the benefits of choice and competition. Car insurance

provides a good example of a sector where consumers frequently

compare prices and switch. But on investigating, the Competition

Commission – now the Competition and Markets Authority – found

that the way in which claims of drivers not at fault were settled by

19 Department of Energy and Climate Change, Quarterly Energy Prices December 2013 (London: HMSO, 2013),

Table 2.2.3

20 Financial Conduct Authority, “Human Face of Regulation”, speech by Martin Wheatley, 2013, http://www.fca.

org.uk/news/speeches/human-face-of-regulation


32

insurance companies was driving up prices.21 Consumer action has

its limits if markets are not functioning properly.

In other areas, helping consumers to get better value means

helping them to do what they already do well in some markets,

where prices and quality are easy to compare. Whilst in some

markets, structural changes may be needed, evidence suggests

that there are more gains to be made from helping more consumers

to become more active.

COMPARING AND SWITCHING

In recent years, Government and regulators have moved to make

switching quicker, easier and more convenient across a range of

markets – energy, financial services and telecoms. This is a positive

step. But the gains from making the process of actually switching

smoother and faster are now close to exhausted. 85–95% of those

who do switch providers – across a whole range of markets – say

that the process was easy. It is only amongst those who have

not switched that there are still perceptions that switching is

difficult.22 Among the middle quintile specifically, the proportion

that perceive switching to be easy is higher than average – across

a range of products including banking and financial products,

energy, telephony and broadband.23

The big gains now are to be found in helping consumers easily

compare the level of savings they can make by switching. In energy

markets, for example, the strongest driver of consumer activity is the

level of savings that consumers expect to make. Even perceptions

of the expected length and difficulty of the switching process itself

21 Competition Commission, “CC seeks to reduce the cost of motor insurance”, 2013, http://www.competition-

commission.org.uk/media-centre/latest-news/2013/Dec/cc-seeks-to-reduce-the-cost-of-motor-insurance

22 Ofcom, The Consumer Experience of 2013 (London: Ofcom, 2013)

23 Which? “Consumer Insight Tracker”, January 2014

RIDERS ON THE STORM

33

have little effect in preventing consumers from moving providers.24

Government and regulators need to ensure that consumers have

access to the right information to make choices, and that it is easy

to use information that does exist to calculate likely savings from

switching.

GIVING PEOPLE ACCESS TO THE RIGHT INFORMATION

A significant area of spending for many middle income families is

on childcare. Websites exist that allow parents to search for nearby

childcare options in their area, and compare Ofsted inspection

reports where they are available. But comparing prices across

different providers is much harder, making it difficult for parents

to compare different types of childcare and decide which is best

for their needs. Adding pricing data to the information that already

exists on information such as Ofsted inspections could make the

process of choosing childcare easier for parents.

In the context of middle income households vulnerable to

increases in mortgage payments in the coming years, a key factor

will be how the Monetary Policy Committee decides to raise rates.

The Governor of the Bank of England has said when rates do rise,

they will do so gradually.25 The extent to which rates rise in tandem

with wage growth will be crucial.

But beyond this, there are other ways to create more choice

and ensure better value for money for consumers. There has been

increased interest from the Bank of England in mortgages that

would allow consumers to fix their rates for a longer period of

24 Miguel Flores and Catherine Waddams Price, “Consumer behaviour in the British retail electricity market”,

(University of East Anglia, 2013)

25 See, for example, Bank of England, “The economics of currency unions”, speech by Mark Carney, 2014,

http://www.bankofengland.co.uk/publications/Pages/speeches/2014/706.aspx


34

time, as is possible in other countries including the US.26 Another

way for households to manage the inevitable rate rises to come is

to compare offers across mortgage providers to get better deals

– whether on fixed or variable rates. There has been confusion

among consumers recently over whether providers offering tracker

mortgages are able to raise their rates in the absence of a Bank

of England rate rise. Lack of transparency and simplicity around

mortgage products will make it much harder for consumers to

manage their finances when rates rise again. It is vital that banks

and regulators ensure that there is sufficient clarity around these

products so that consumers can make informed choices. Measures

such as the implementation of the Financial Conduct Authority’s

Mortgage Market Review will help with this.

LIBERATING DATA

There is also much more to do on helping consumers easily manage

their spending, compare products and pick the best deals using

data that already exists.

Innovation in banking and use of financial data can play a

valuable role here. Mobile banking innovators have already started

to explore ways of using banking data to better help consumers

understand and analyse their outgoings using transactional

information on how they spend their money.27 The midata initiative,

which is designed to encourage companies to release consumer

data such as usage statistics and the development of apps that

make it easier to process this usage data and compare prices is also

very promising step.28 Building on midata, the proposal to include

26 Treasury Committee, “Bank of England Financial Stability report hearings”, 2013, http://www.parliament.uk/

business/committees/committees-a-z/commons-select/treasury-committee/inquiries1/parliament-2010/

bank-of-england-financial-stability-report-hearings/

27 Cormac Hollingsworth and Emran Mian, Branching out (London: Social Market Foundation, 2014)

28 Department for Business, Innovation and Skills, “Providing better information and protection for consumers”,

2014 https://www.gov.uk/government/policies/providing-better-information-and-protection-for-consumers/

supporting-pages/personal-data

RIDERS ON THE STORM

35

QR codes on energy bills could help dramatically increase the ease

and convenience of comparing providers using the customer’s

own usage data.29 Even larger benefits could come from services

that analyse data across the market to show individual consumers

how their spending compares to the average. For example, in

energy markets, knowing how much more they are paying per unit

consumed compared to other households in their local area could

make the potential savings much more visible.

In some cases, the market will develop these new services

and applications itself. But Government needs to remain aware of

the fact that in other cases, relying on voluntary participation by

companies will not be enough. The case of QR codes on energy bills

is a recent example of where relying on voluntary action by energy

companies has not worked, and Government is now consulting on

modifying energy company licences to have QR codes included as

part of energy bills.

There is likely to be a first-mover disadvantage from voluntarily

participating in schemes such as the provision of QR codes and the

midata scheme more generally. This is because the first company

to release usage data to its customers is likely to be subject to

more switching away to its competitors, creating a commercial

disadvantage. Given this dynamic, there is a strong rationale to

ensure that major players in the same industry act at the same

time, whether this is achieved through voluntary agreement or

compulsion.

Sectors where this scheme has the most beneficial applications

are, firstly, ones where consumers tend to have ongoing

relationships with suppliers, and therefore where it is highly likely

that data is already being collected; and secondly, where data on

29 Department of Energy and Climate Change, “QR codes on energy bills put consumers in control”, 2014,

https://www.gov.uk/government/news/qr-codes-on-energy-bills-put-consumers-in-control


36

consumption or usage is needed to compare offers across different

providers effectively. At the moment, Government is looking

to ensure that customers of current account providers, mobile

phone companies and energy providers benefit from greater

access to data on their usage. The Treasury recently announced

that the major current account providers will give customers their

account data in a simple, standardised format that can be used in

comparison sites.30

But there are other products and services that the midata

scheme could be expanded to. In telecoms, there are likely to

be benefits from expanding beyond mobile phone companies.

As mobile and landlines have increasingly become substitute

products,31 it would be beneficial for consumers to be able to

compare usage and tariffs across both the landline and mobile

calls they make. Broadband is another service where consumers – if

armed with better information on how much data they typically

use – could make a more informed decision as to which broadband

package and speed is best for them. Within the financial sector,

information on a consumer’s typical incomings and outgoings

could help consumers better understand what types of savings

products are right for them – for example instant access versus

notice accounts and regular savings accounts versus those where

the rate does not depend on how often money is deposited.

Outside the regulated sector, supermarkets, especially those with

loyalty cards, are likely to have a wealth of data that could better

help consumers manage grocery spending.

JUST PUSHING DATA IS NOT ENOUGH

Policymakers should also learn the lessons from price comparison

websites, which have made it easier to compare products, but

30 HM Treasury, Budget 2014 (London: HMSO, 2014)

31 Ofcom, International communications market report (London: Ofcom, 2013)

RIDERS ON THE STORM

37

do have limitations. Firstly, older consumers are less likely to use

these sites.32 Secondly, there have been concerns about how

well these sites capture data on actual pricing – as opposed to

offering comparisons based on headlines prices only, and whether

in some cases conflicts of interest may mean that sites do not

provide fully objective comparisons.33 And thirdly, the effect of

actively prompting consumers to compare prices should not be

underestimated: consumers say that renewal notices for services

such as car insurance act as a trigger for them to look at what

other providers are offering.34 Switching rates for car insurance are

36% – triple the rate for energy and seven times the rate for bank

accounts.

Innovative applications are likely to play a significant role in

making comparisons easier, and they may well also develop to act

as prompts for consumers, for example by keeping track of usage

and alerting consumers when deals that better fit their needs

come onto the market. But where this does not happen, other

measures need to be considered by Government, regulators and

industry, proceeding through voluntary agreement, or compulsion

if necessary. The model of renewal notices in car insurance offers a

guide to what could work well for other sectors. Receiving a letter

once a year as a prompt, especially if combined with an easy to use

system of comparing offers using the consumer’s own usage data

could be very powerful.

Given competing pressures on public spending, providing

substantial help to middle income households through tax

reductions or more cash transfers is difficult. As recent analysis has

shown, Government is still less than half way through its planned

32 Consumer Futures, Price comparison websites: consumer perceptions and experiences (London: Consumer

Futures, 2013)

33 Financial Conduct Authority, “The FCA launches review into price comparison websites”, 2013, http://www.

fca.org.uk/news/the-fca-launches-review-into-price-comparison-websites

34 Consumer Futures, Price comparison websites: consumer perceptions and experiences


38

spending cuts.35 But middle income households have shown

remarkable resilience so far, demonstrating that under the right

circumstances, and with the right information to hand they can

manage their finances and cope with being under pressure. Food

spending is one example, where they have shopped around to

keep down costs in spite of rising prices. By equipping consumers

to negotiate markets where it is currently harder to secure better

value for money, middle income households can be helped to

better manage their finances in the future. Making this happen will

require action from Government, regulators and industry.

35 Institute for Fiscal Studies, “Still not halfway there yet on planned spending cuts”, 2014, http://www.ifs.org.

uk/publications/7086

RIDERS ON THE STORM

39

ANNEX: METHODOLOGY

OVERVIEW

We use longitudinal data from the British Household Panel

Survey (BHPS) and Understanding Society (US)36 to follow specific

households over time. We look at how households’ position in

the income distribution has changed, and focus on where the

households currently in the middle and fourth quintiles have come

from.

Our definition of quintiles is based on original income before

taxes and benefits to understand underlying trends before

government transfers, however, we take into account gross income

including benefits when looking at the spending squeeze. We

look at income, employment, spending patterns and leisure as

indicators of how households have fared.

There are some areas where there is insufficient information

in the BHPS and US surveys. We therefore supplement the BHPS/

US analysis with additional cross-sectional analysis from the Family

Resources Survey (FRS) and the Living Costs and Food Survey (LCFS),

for data on childcare, leisure and energy. Because these datasets

are cross-sectional, any changes over time are based on quintiles as

defined in specific years, and do not follow households over time.

Our analysis is based on data from the latest waves of each of these

surveys, 2011–12 (BHPS/US and FRS), and 2011 (LCFS). We also use

the FRS to show a snapshot of how the income distribution has

changed between 2007–08 and 2011–12.

For inflation-adjusted comparisons, we use the Consumer Price

Index.

36 Understanding Society is an initiative by the Economic and Social Research Council, with scientific

leadership by the Institute for Social and Economic Research, University of Essex, and survey delivery by the

National Centre for Social Research.


40

ANALYSIS OF UNDERSTANDING SOCIETY AND THE BRITISH HOUSEHOLD PANEL STUDY

The main analysis presented in this report is based on data from

two longitudinal datasets, British Household Panel Survey (BHPS)

and Understanding Society (US). Beginning in 1991, the BHPS was a

UK representative survey that followed the same set of individuals

and households for eighteen consecutive years. The final wave of

BHPS, Wave 18, was completed in April 2009.

Launched in 2010 as the successor to the BHPS, US is a

UK representative longitudinal dataset that surveys the same

individuals and households over time. Understanding Society

incorporated almost 6,700 individuals that had previously been

in the BHPS sample. In total, the latest wave of Understanding

Society (Wave 3) contains data from 49,739 individuals and

27,783 households, and was collected between January 2011 and

December 2012.

The first stage of our analysis focused on this sample of

households from US. Of this sample, we excluded households that

had at least one member of pensionable age (defined as 60 for

women and 65 for men), as our analysis is focused on the working

age population only. Accordingly, approximately 33% of the original

household dataset was excluded and the resultant dataset was

composed of 34,656 individuals from 18,518 households.

When we conducted our analysis, derived data on original

income before taxes and benefits was unavailable. We therefore

constructed our own original income variable by subtracting

household monthly benefit income from the state from monthly

gross household income. We used this original income variable to

construct our income distribution and assign income quintiles to

the households in our dataset. We then analysed household and

individual characteristics within these groups to produce top-level

RIDERS ON THE STORM

41

statistics on what middle income households looked like in 2011–12

and how they differed from households in other quintiles.

For the second stage of our analysis, we tracked households

over time, comparing their situation in 2007–08 against that in

2011–12. Specifically, we matched households from Wave 17 of

the BHPS to households in Wave 3 of US. We excluded households

that split. We also excluded any households that contained an

individual of pension age in either one of the two waves to avoid

picking up any trends related to entering retirement. The resultant

sample consisted of 3,166 successfully matched households with

5,931 individuals.

As with the US dataset, we allocated households a quintile

based on the income distribution in the BHPS 2007–08 dataset.

This meant that each of our matched households had two quintiles

associated with them – one for 2007–08 and one for 2011–12.

In analysing characteristics across the two years, we made the

following decisions to allow comparability:

• Food: In contrast to US, BHPS denotes household expenditure

on food in ranges. We convert expenditure ranges to

actual monetary values by taking the midpoint of each

food expenditure band. To make food expenditure in BHPS

comparable to US, we derived variables in US and BHPS that

captured monthly expenditure on food and groceries as well

as food bought outside the home. This required summing

the two food variables in US, monthly household expenditure

on groceries and monthly household expenditure on

food bought outside the home (including restaurants and

takeaways) respectively. For BHPS, we aggregated monthly

expenditure on food consumed in the home and household

expenditure on food and groceries including takeaways.

• Energy: In both BHPS and US, household expenditure on


42

gas, electricity and other fuels are separate variables. We

generated a new variable that aggregates each component of

energy expenditure to obtain an estimate of household total

expenditure on energy in the previous 12 months.

• Mortgage and rent: In both US and BHPS, mortgage paid is

a variable in monthly terms whilst rent paid is a variable that

covers a range of time periods. Accordingly, we transformed

this rent variable to monthly terms using variables detailing

amount of rent paid in last instalment and weeks covered by

rent paid. We then generated a new variable that captures

both monthly mortgage and rent paid in last month.

ANALYSIS OF THE FAMILY RESOURCES SURVEY

The Family Resources Survey (FRS) is a cross-sectional dataset used

for the analysis of household incomes, inequality and household

expenditure. For the analysis in this report we used data for the

years 2007–08 and 2011–12 for childcare expenditure patterns and

for the analysis of general patterns in the change in the income

distribution between 2007–08 and 2011–12.

The total sample size of the survey is 20,759 households for

the year 2011–12 and 24,977 households for the year 2007–08. We

excluded households that had at least one member of pensionable

age (defined as 60 for women and 65 for men), as our analysis is

focused on the working age population only. This left a sample of

16300 households for 2007 and 13581 households for 2011.

We used the FRS to show how the UK’s income distribution

profile changed between 2007–08 and 2011–12. To do this, we

constructed household original income (before taxes and benefits)

to assign quintiles. We did this by aggregating gross income from

employment, self-employment, investment income, and other

non-benefit income.

RIDERS ON THE STORM

43

We also used the FRS for data on childcare. About 55% of

households did not contain any children (stable for both waves of

the FRS we looked at), so our analysis of childcare usage and childcare

expenditure was based on about 7,340 and 6,200 households in

2007–08 and 2011–12, respectively. We then computed the average

childcare cost by quintile and number of hours used, both of free

of charge childcare and of paid for childcare, per child for different

age groups. This calculation was based on summing expenditure

across different types of childcare for each child within the dataset,

as well as the number of hours used of both informal and informal

childcare.

ANALYSIS OF THE LIVING COSTS AND FOOD SURVEY

The Living Costs and Food Survey is a cross-sectional dataset

collecting information on the expenditures of households in

the UK. It has been collected since 2008 as a replacement to the

Expenditure and Food Survey.37 The survey had a sample size of

about 5700 households in 2011, and about 6100 in 2007. For this

report we have used data on expenditure on leisure goods and

services, and expenditures for internet and fuel.

As with the FRS, we excluded households with members of

pension age, leaving around 4,000 households in each year we

analysed. We grouped households into income quintiles using the

same methodology as for the FRS, first constructing an original

income variable including wages and earnings income, income

from self-employment, investment income and other sources, plus

income from private pensions.

Analysis of the expenditure on leisure goods and services and

sub-items within this group were then performed by computing

weighted averages across quintile groups.

37 For more information see UK Data Service, http://discover.ukdataservice.ac.uk/series/?sn=2000028.

ISBN: 978-1-904899-89-1£10.00

SOCIAL MARKET FOUNDATION11 Tufton Street | Westminster | London SW1P 3QBPhone: 020 7222 7060 | Fax: 020 7222 0310 www.smf.co.uk

THE SO

CIA

L MA

RK

ET FOU

ND

ATIO

NR

IDER

S ON

THE STO

RM

: BR

ITAIN

’S MID

DLE IN

CO

ME H

OU

SEHO

LDS SIN

CE 2007

N

IDA

BR

OU

GH

TON

, ON

YIN

YE EZ

EYI, C

LAU

DIA

HU

PKA

U

In the years since the financial crisis, working age households have seen wage stagnation and rises in the cost of living, prompting widespread focus on the so-called “squeezed middle”. This report looks at the reality of middle income households coming out of the downturn – using the British Household Panel Study and Understanding Society to track specific households over time.

Middle income households have shown remarkable resilience. Many have taken on more work and increased their incomes. There is a surprising amount of movement across the income distribution, with many climbing up. Middle income households have also shopped around and cut back to reduce the effect of rising prices.

But for many of these households, the squeeze may be yet to come when mortgage rates rise again. And in some markets it is much harder to compare products and secure value for money. With competing pressures on public spending, these households can nevertheless be helped by supporting them to do what they already do well – shopping around, comparing prices and switching.

riders on the storm - home, the social market … · riders on the storm contents acknowledgements...

Documents